dissenting.
The majority concludes that the district court did not abuse its discretion in denying jurisdictional discovery. I respectfully disagree with that conclusion, and there*628fore dissent. My disagreement with the majority stems in large part from the majority’s treatment of Calder v. Jones, 465 U.S. 783, 104 S.Ct. 1482, 79 L.Ed.2d 804 (1984). In assessing whether the government’s tort claims arise out of or relate to SAB’s contacts with the forum, the majority states that, “since Calder’s ‘effects’ test is relevant only to the purposeful availment prong, it cannot be used to strengthen the government’s relatedness showing.” That reasoning cannot be squared with Calder’s holding that jurisdiction can be “based on” the in-forum effects of the defendant’s out-of-forum activity. 465 U.S. at 787, 104 S.Ct. 1482. Under Calder, those effects are jurisdictional contacts in their own right, relevant to the relatedness requirement.
Although I agree with the majority that the government has not yet made out a prima facie case for specific jurisdiction, I believe that the government’s effects argument creates a “colorable” case for specific jurisdiction with respect to its tort claims against SAB. Accordingly, I conclude that the district court abused its discretion in summarily denying the government’s request for jurisdictional discovery on the ground that the government’s case for personal jurisdiction is “bootless.”
I.
My disagreement with the majority over the import of Calder leads me to a different view on the question of jurisdictional discovery. Thus, before turning to the discovery question, I first must address Calder itself, and its implications for the government’s case for specific jurisdiction.
A. The Jurisdictional Relevance of Effects
The dispute in Calder arose out of an allegedly libelous article published in the National Enquirer about Shirley Jones, a well-known California entertainer. Jones sued the Enquirer, Ian Calder, its president and editor, and John South, the reporter who wrote the offending article. Calder and South were both Florida residents, and it was undisputed that the article had been written, researched, and edited in Florida. Indeed, Calder never even called California in connection with the article: “all of his acts with reference to [the Jones] article apparently were performed in Florida.” Jones v. Calder, 138 Cal.App.3d 128, 187 CaLRptr. 825, 829 (1982).
The California Court of Appeal concluded that “[t]he fact that the actions causing the effects in California were performed outside the State did not prevent the State from asserting jurisdiction over a cause of action arising out of those effects.” Calder, 465 U.S. at 787, 104 S.Ct. 1482. The Supreme Court agreed, noting its “approval of the ‘effects’ test employed by the California court.” Id. at 787 n. 6, 104 S.Ct. 1482. That test was drawn from § 37 of the Restatement (Second) of Conflicts of Laws, which provides:
A state has power to exercise judicial jurisdiction over an individual who causes effects in the state by an act done elsewhere with respect to any cause of action arising from these effects unless the nature of the effects and of the individual’s relationship to the state make the exercise of jurisdiction unreasonable.
As the language of the Restatement test suggests, its elements mirror those of our traditional specific jurisdiction inquiry. The first clause, authorizing jurisdiction over “one who causes effects in the state by an act done elsewhere,” establishes that in-forum effects are relevant contacts for the jurisdictional analysis. The second clause then limits the exercise of jurisdiction to cases in which there is a sufficient nexus between the defendant’s forum contacts (here, the in-forum effects) and the *629plaintiffs cause of action. That clause correlates to the relatedness requirement for specific jurisdiction, which is satisfied when the plaintiffs cause of action either “aris[es] out of or relate[s] to the defendant’s contacts with the forum.” Helicopteros Nacionales de Colombia, S.A. v. Hall, 466 U.S. 408, 414 n. 8, 104 S.Ct. 1868, 80 L.Ed.2d 404 (1984) (emphasis added). We have said that “we think it significant that the constitutional catchphrase is disjunctive in nature, referring to suits arising out of or relating to in-forum activities. We believe that this added language portends added flexibility and signals a relaxation of the applicable standard.” Ticketmaster-NY, Inc. v. Alioto, 26 F.3d 201, 206 (1st Cir.1994) (citations and internal quotation marks omitted). There is no reason to depart from our usual understanding of the relatedness inquiry in this case. Therefore, although the Restatement uses “arising out of’ language to describe its relatedness requirement, the requirement also can be satisfied by a showing that the plaintiffs cause of action “relates to” the in-forum effects of the defendant’s activity.
The final clause of the effects test adds a proviso, forbidding effects-based jurisdiction in cases where “the nature of the effects and of the individual’s relationship to the [forum] make the exercise of jurisdiction unreasonable.” Restatement (Second) of Conflict of Laws, § 37. Prior to Calder, the Supreme Court had explained that the general “reasonableness” inquiry mandated by the effects test overlaps in large part with the purposeful availment inquiry. See Kulko v. Superior Court, 436 U.S. 84, 98 S.Ct. 1690, 56 L.Ed.2d 132 (1978). Thus, effects-based jurisdiction is “unreasonable” under the Restatement test where the defendant has not intentionally reached out to the forum state in some way, so that he or she reasonably could anticipate being haled into court there. See id. at 96-98, 98 S.Ct. 1690.
The circumstances of Calder easily satisfied the first two clauses of the Restatement test. The article, written by defendants Calder and South in Florida, had caused harmful effects in the forum state; as the Court observed, “the brunt of the harm [to Jones], in terms both of [her] emotional distress and the injury to her professional reputation, was suffered in California.” Calder, 465 U.S. at 789, 104 S.Ct. 1482. Jones’s cause of action arose out of those effects. See id. at 787, 104 S.Ct. 1482. Thus, as the majority explains, since the in-forum effects of the Calder defendants’ actions “were clearly related to the plaintiffs defamation suit, ... the Supreme Court did not need to address the relatedness prong before proceeding to the purposeful availment inquiry.”
The Court began that inquiry by distinguishing the defendants’ situation from that of a hypothetical welder who works on a boiler in Florida that later explodes in California. See id. at 789, 104 S.Ct. 1482. The welder obviously can “foresee” that the boiler might make its way to California and cause harmful effects there. Id. Yet, the Court observed, it may well be unfair to subject the welder to jurisdiction in California when he “has no control over and derives no benefit from his employer’s sales in that distant State.” Id.
Unlike the unfortunate welder, Calder and South were “not charged with mere untargeted negligence.” Id. Rather, the Court emphasized, “their intentional, and allegedly tortious, actions were expressly aimed at the forum state.” Id. The allegedly libelous story “concerned the California activities of a California resident” whose “career was centered in California.” Id. at 788-89, 104 S.Ct. 1482. Moreover, the defendants knew the article “would have a potentially devastating impact” on Jones, and that she would suffer “the *630brunt of the injury” in California, where she lived and worked. Id. at 789-90, 104 S.Ct. 1482. In short, California was “the focal point both of the story and of the harm suffered.” Id. at 789, 104 S.Ct. 1482. Thus, the defendants “must ‘reasonably anticipate being haled into court there,’ ” id. at 790, 104 S.Ct. 1482 (quoting WorldWide Volkswagen Corp. v. Woodson, 444 U.S. 286, 297, 100 S.Ct. 559, 62 L.Ed.2d 490 (1980)), and jurisdiction reasonably could be “based on the ‘effects’ of [defendants’] Florida conduct in California,” id. at 789, 104 S.Ct. 1482.
B. The Majority’s Reading: Calder and Related Contacts
Given Calder’s focus on the reasonableness of exercising jurisdiction on the basis of effects, it is easy to understand the majority’s assertion that Calder “is a gauge for purposeful availment.” When a plaintiff seeks to base jurisdiction on the in-forum effects of the defendant’s activity elsewhere, the case likely will turn on such questions as whether the defendant’s allegedly tortious conduct was intentionally and “expressly aimed” at the forum state, and whether the “brunt of the harm” was felt there. Calder, 465 U.S. at 789, 104 S.Ct. 1482. Those inquiries properly fall under the purposeful availment prong because they are designed to determine whether the defendant intentionally i'eached out to cause harm in the forum state.8
Contrary to the conclusion of the majority, however, it does not follow that Calder “is relevant only to the purposeful availment prong [and so] cannot be used to strengthen the government’s relatedness showing.” As I have explained, the Restatement “effects” test approved in Calder includes a relatedness element. It permits a state to exercise effects-based jurisdiction only when the plaintiffs claims arise out of or relate to the in-forum effects of the defendant’s acts. See Restatement (Second) of Conflict of Laws, § 37; Calder, 465 U.S. at 787, 104 S.Ct. 1482 (noting that effects-based jurisdiction was proper where Jones’s claims arose out of the California effects of the defendants’ actions).
The majority does not suggest that the government’s claims against SAB are not related to the in-forum effects of SAB’s allegedly tortious activity. Accordingly, when it says that the effects test “cannot be used to strengthen the government’s relatedness showing,” the majority must mean that, under Calder, the in-forum effects are not jurisdictional contacts themselves, but merely additional evidence that the defendants acted purposefully. Based on that interpretation of Calder, the majority states that the relatedness inquiry can be satisfied only when “the defendant’s in-forum conduct caused the injury or gave rise to the cause of action.”
However, the effects test adopted in Calder explicitly authorizes jurisdiction based on the in-forum effects of “an act done elsewhere.” Restatement (Second) of Conflict of Laws, § 37. Those effects are relevant jurisdictional contacts, apart from *631any link between the plaintiffs tort claims and the defendant’s “in-forum conduct.” Thus, in Calder, the Court did not rely on the presence of physical, mail, or telephone contacts between the defendants and the forum. Instead, it held that jurisdiction was proper “based on ‘effects’ of [defendants’] Florida conduct in California.” Calder, 465 U.S. at 789, 104 S.Ct. 1482; see also Hugel v. McNell, 886 F.2d 1, 4 (1st Cir.1989) (explaining that, under Calder, “[t]he knowledge that the major impact of the injury would be felt in the forum State constitutes a purposeful contact or substantial connection whereby the intentional tortfeasor could reasonably expect to be haled into the forum State’s courts to defend his actions”); Haisten v. Grass Valley Med. Reimbursement Fund, Ltd., 784 F.2d 1392, 1397 (9th Cir.1986) (noting that in Calder “the Court ... allowed the exercise of jurisdiction over a defendant whose only ‘contact’ with the forum state [was] the ‘purposeful direction’ of & foreign act having effect in the forum state” (first emphasis added)). It is difficult to understand how jurisdiction could have been permissible in those circumstances were the in-forum effects of acts done elsewhere not themselves contacts.
Indeed, the majority’s methodology would seem to compel a result contrary to that reached in Calder. On the majority’s understanding, “the effects test ... is to be applied only after the relatedness prong has been satisfied.” That creates a quandary for the plaintiff whose cause of action arises out of or relates to the in-forum effects of out-of-forum activity. If those effects are off-limits during the relatedness inquiry, and if that inquiry must be completed before the effects can be taken into account under the purposeful availment analysis, then the plaintiff never will be able to establish jurisdiction “based on” those effects. Calder, therefore, is a dead letter. The only cases in which in-forum effects could be considered are those in which jurisdiction might just as easily be based on some other forum contacts.
The majority offers two bases for its reading of Calder. First, it emphasizes that in Noonan v. Winston Co., 135 F.3d 85, 90 (1st Cir.1998), we said that Calder “adopted an effects test for determining purposeful availment in the context of defamation cases.” It is important to see that statement in context:
The decisive due process issue in this [defamation] case is whether the defendants’ activities satisfy the purposeful availment requirement. Plaintiffs correctly draw our attention to Calder v. Jones, in which the Supreme Court adopted an effects test for determining purposeful availment in the context of defamation cases.
Id. (internal citation omitted). Noonan cannot bear the weight the majority gives it. Calder did establish a test for determining purposeful availment in defamation cases. The majority’s reading depends on the entirely different point that Calder did not also establish that jurisdiction can be based on the in-forum effects of out-of-forum activity when such effects relate or give rise to the cause of action. Noonan did not discuss relatedness at all, and so provides no support for the majority’s restrictive interpretation of Calder.
Second, the majority points out that “we have wrestled before with [the] issue of whether the in-forum effects of extra-forum activities suffice to constitute minimum contacts and have found in the negative.” Mass. Sch. of Law v. Amer. Bar Ass’n, 142 F.3d 26, 35-36 (1st Cir.1998). In further support of that point, the majority cites Kowalski v. Doherty, Wallace, Pillsbury & Murphy, 787 F.2d 7 (1st Cir.1986), and Sawtelle v. Farrell, 70 F.3d 1381 (1st Cir.1995), in which we held that New Hampshire could not exercise jurisdiction over foreign law firms based on *632allegedly negligent acts committed outside the state.9 We discussed those cases in Massachusetts School of Law, and concluded that, “[j]ust as the New Hampshire effects of [out-of-state] negligence, without more, could not sustain an action in New Hampshire against the negligent actor, see Kowalski, 787 F.2d at 11, so too the Massachusetts effects of the [defendants’] [out-of-state] actions, without more, fail to sustain an action in a Massachusetts court.” 142 F.3d at 36 (also citing Sawtelle, 70 F.3d at 1394).
We did not mention Calder—much less rely on it — in Massachusetts School of Law, Kowalski, or Sawtelle. Nevertheless, our holdings in those cases are consistent with the effects test that I have described. Under Colder, in order for jurisdiction to be based solely on the in-forum effects of the defendant’s activity, the plaintiff must show that the defendant acted “for the very purpose” of causing harmful effects in the forum. Lake v. Lake, 817 F.2d 1416, 1422 (9th Cir.1987); Restatement (Second) of Conflict of Laws, § 37 cmt. e (“When the act was done with the intention of causing the particular effects in the state, the state is likely to have judicial jurisdiction though the defendant had no other contact with the state.”). No such showing was made (or even attempted) in Massachusetts School of Law, Sawtelle, and Kowalski. In those cases, therefore, effects-based jurisdiction would have been “unreasonable” under the Restatement test, not because the in-forum effects were not contacts, but because the “nature of the effects” was such that jurisdiction could not rest on them alone. Restatement (Second) of Conflict of Laws, § 37; see also Calder, 465 U.S. at 789-89, 104 S.Ct. 1482; Kulko, 436 U.S. at 96-97, 98 S.Ct. 1690.
To be sure, in-forum effects that lack the requisite intentionality are still jurisdictional contacts that must be taken into account in the overall analysis. Calder compels that conclusion, and our cases do not suggest otherwise. But other contacts between the defendant, the forum, and the litigation are necessary in order to render the exercise of jurisdiction reasonable. See Restatement (Second) of Conflict of Laws, § 37 cmt. e (“The fact that the effect in the [forum] was ... foreseeable will not itself suffice to give the [forum] judicial jurisdiction over the defendant.”); Panda Brandywine Corp. v. Potomac Elec. Power Co., 253 F.3d 865, 869 (5th Cir.2001) (explaining that “the effects of an alleged intentional tort are to be assessed as part of the analysis of the defendant’s relevant contacts with the forum” (internal quotation marks omitted)).
Here, the majority does not dispute that SAB’s actions caused harmful effects — the loss of money' — in the United States. Nor does it dispute that those harmful effects are related to the government’s claims of wrongful conversion and unjust enrichment. The crucial question, therefore, is whether SAB’s actions satisfy the purposeful availment inquiry; that is, whether SAB “expressly aimed” its allegedly tor-tious activity at the United States with the knowledge that “the brunt of the harm” would be felt there. Calder, 465 U.S. at 789, 104 S.Ct. 1482.
C. Purposeful Availment
We have said that Colder imposes a two-part test for purposeful availment, requiring a plaintiff to show (1) that it felt the injurious effects of a defendant’s tortious act in the forum, and (2) that the defendant’s act was “calculated to cause injury” *633to the plaintiff there. Noonan, 135 F.3d at 90 (citing Calder, 465 U.S. at 791, 104 S.Ct. 1482). The government easily satisfies the first prong. The loss of the forfeit $7 million to the United States government as a result of SAB’s alleged conversion and unjust enrichment necessarily had injurious effects that were felt in the United States. In Swiss II, we concluded that the “legal injuries] occasioned by” the torts of conversion and unjust enrichment occurred in Antigua, where the conversion and enrichment took place. United States v. Swiss Am. Bank, Ltd., 191 F.3d 30, 37 (1st Cir.1999). Nevertheless, we acknowledged that, “upon the occurrence of the alleged conversion and the consequent unjust enrichment, the United States felt the effects of a tortious injury in the [United States].” Id. at 38.
The majority suggests that the fact that the government’s injury occurred in Antigua distinguishes this case from Calder. Cf. Kowalski, 787 F.2d at 11 (distinguishing between injury and effects for purposes of the New Hampshire long-arm statute, which requires that the plaintiffs injury occur in the forum); Sawtelle, 70 F.3d at 1390 (explaining that, in Kowalski, “we rejected the plaintiffs contention that, because the ‘effects’ of the [defendant law] firm’s negligence were felt in New Hampshire, the law firm had caused an injury there by conduct directed at that forum .... The injury, if any, occurred in Massachusetts”). Yet here, as in Calder, the plaintiff suffered “the brunt of the harm” in the forum. Calder, 465 U.S. at 789, 104 S.Ct. 1482. That similarity suggests that the outcome of our jurisdictional analysis should not be different in this case simply because the injury caused by libel is deemed to occur wherever the libelous material is circulated, while the injury of conversion is deemed to occur where the conversion took place. Such formalistic distinctions can be helpful in cases like Swiss II, where the applicable state long-arm statute requires an in-forum injury as a prerequisite to jurisdiction. See Swiss II, 191 F.3d at 38 (applying § 3(d) of the Massachusetts long arm statute, which authorizes jurisdiction over one who, inter alia, causes “tortious injury in this commonwealth”); see also Kowalski, 787 F.2d at 11 (applying similar New Hampshire statute). Our inquiry here is not so rigidly confined, where strict rules give way to “traditional notions of fair play and substantial justice.” Int’l Shoe Co. v. Washington, 326 U.S. 310, 316, 66 S.Ct. 154, 90 L.Ed. 95 (1945) (internal quotation marks omitted). Given the flexible nature of our due process analysis, we should hesitate before adopting a bright-line rule that in-forum effects do not constitute jurisdictional contacts unless they also can be deemed an “injury.”
That is not to say that the situs of the plaintiffs injury is irrelevant to the jurisdictional analysis. In cases where the injury occurred outside the forum, the plaintiff may find it difficult to satisfy the second prong of the Calder test, which requires a showing that the defendant’s act was “calculated” to cause the harmful effects in the forum. That inquiry is designed to determine whether the nature of the effects is such that jurisdiction reasonably can be based on them alone, and it is here that the government’s prima facie case for jurisdiction falters. The government argues that “SAB knew that its intentional conduct in Antigua would cause injury to the United States government.” That is not enough. The government must show that SAB’s actions were “expressly aimed” at the United States as a forum. Calder, 465 U.S. at 789, 104 S.Ct. 1482 (distinguishing the case of the negligent welder); Wien Air Alaska, Inc. v. Brandt, 195 F.3d 208, 212 (5th Cir.1999) (“Foreseeable injury alone is not sufficient to confer specific jurisdiction, absent the direction of specific acts toward the fo*634rum.”); cf. Burger King, 471 U.S. at 474, 105 S.Ct. 2174 (“Although it has been argued that foreseeability of causing injury in another State should be sufficient to establish [minimum] contacts there ..., the Court has consistently held that this kind of foreseeability is not a ‘sufficient benchmark’ for exercising personal jurisdiction.” (footnote omitted) (quoting World-Wide Volkswagen, 444 U.S. at 295, 100 S.Ct. 559)).
The government argues that SAB’s intentional defiance of the preliminary forfeiture order issued by the district court constitutes such express aiming. The forfeiture order identified the forfeited property as “funds which were deposited into the Swiss American Bank, Ltd., and the Swiss American National Bank in St. Johns, Antigua during the time period September 1985 through June 23, 1987.” It is undisputed that SAB was aware of the order, and responded by writing to the district court to inform it that the Anti-guan government had frozen Fitzgerald’s accounts. However, the fact that SAB had notice that the money it took for itself belonged to the United States government does not, in itself, make the United States as a forum the focal point of SAB’s allegedly tortious activity. As the Third Circuit has observed, Calder did not “carve out a special intentional torts exception to the traditional specific jurisdictional analysis, so that a plaintiff could always sue in his or her home state.” IMO Indus., Inc. v. Kiekert AG, 155 F.3d 254, 265 (3d Cir.1998). Therefore, it cannot be enough that the defendant knew when it acted that its victim lived in the forum state. See id. (“Simply asserting that the defendant knew that the plaintiffs principal place of business was located in the forum would be insufficient in itself to meet [the ‘expressly aimed’] requirement.”); accord Southmark Corp. v. Life Investors, Inc., 851 F.2d 763, 773 (5th Cir.1988) (concluding that the location of the plaintiff's principal place of business in the forum was a “mere fortuity,” insufficient to show that the defendant expressly aimed its actions at the forum); ESAB Group, Inc. v. Centricut, Inc., 126 F.3d 617, 625 (4th Cir.1997) (finding no jurisdiction where the defendant knew that its acquisition of the plaintiffs trade secrets would result in lower sales for the plaintiff, but did not “manifest behavior intentionally targeted at and focused on” the forum state). Something more is needed to show that SAB’s actions were “purposefully directed” or “expressly aimed” at the United States.
Because the government has not demonstrated that SAB’s actions were “intentionally targeted at and focused on the forum,” IMO Indus., 155 F.3d at 263, the in-forum effects of those actions do not provide a sufficient basis for the exercise of jurisdiction. Although those effects qualify as a relevant (and, as I have explained, related) contact between SAB and the United States as a forum, that contact is too “attenuated” to satisfy the requirement of purposeful availment. Burger King, 471 U.S. at 475, 105 S.Ct. 2174 (internal quotation marks omitted). In the words of the Restatement, “the nature of the effects and of the [defendant’s] relationship to the [forum] make the exercise of [effeets-based] jurisdiction unreasonable.” Restatement (Second) of Conflict of Laws, § 37. Thus, the government must demonstrate that SAB had other contacts with the forum “such that the maintenance of the suit does not offend ‘traditional notions of fair play and substantial justice.’ ” Int’l Shoe, 326 U.S. at 316, 66 S.Ct. 154 (quoting Milliken v. Meyer, 311 U.S. 457, 463, 61 S.Ct. 339, 85 L.Ed. 278 (1940)).
D. The Gestalt Factors
Although I have concluded that the in-forum effects of SAB’s activity lack the requisite purposefulness to support juris*635diction on their own, my inquiry does not end there. The Supreme Court has laid out five criteria for assessing the overall reasonableness of an exercise of personal jurisdiction. See Burger King, 471 U.S. at 476-77, 105 S.Ct. 2174. In close cases, those criteria — which we have termed the “gestalt factors,” see Foster-Miller, Inc. v. Babcock & Wilcox Canada, 46 F.3d 138, 150 (1st Cir.1995)—“may tip the constitutional balance.” Nowak v. Tak How Inv., Inc., 94 F.3d 708, 717 (1st Cir.1996); accord Burger King, 471 U.S. at 477, 105 S.Ct. 2174 (explaining that gestalt factors “sometimes serve to establish the reasonableness of jurisdiction upon a lesser showing of minimum contacts than would otherwise be required”). Even if they do not alter the constitutional balance, the gestalt factors can be important in determining whether the plaintiffs jurisdictional showing is “colorable” enough to support a request for jurisdictional discovery. Therefore, the jurisdictional inquiry is incomplete in this case without consideration of the gestalt factors.
Those factors are “the plaintiffs interest in obtaining convenient and effective relief; the burden imposed upon the defendant by requiring it to appear; the forum’s adjudicatory interest; the [forum] judicial system’s interest in the place of adjudication; and the common interest of all affected sovereigns ... in promoting substantive social policies.” Donatelli v. Nat’l Hockey League, 893 F.2d 459, 465 (1st Cir.1990). We refer to them as the “gestalt” factors “because, in any given ease, they may neither be amenable to mechanical application nor be capable of producing an open-and-shut result. Their primary function is simply to ... put[ ] into sharper perspective the reasonableness and fundamental fairness of exercising jurisdiction.” Foster-Miller, Inc., 46 F.3d at 150.
In assessing the burden of appearance on the defendant, we have considered whether the defendant does business with the forum, Nowak, 94 F.3d at 718, and the distance between the defendant’s place of business and the forum, Ticketmaster-NY, Inc., 26 F.3d at 210. As the majority has explained, the record does not show that SAB does business in the United States. In addition, the distance from Antigua to the United States is “appreciable.” Id. For these reasons, the burden on SAB of litigating in a United States district court in Massachusetts is a relatively heavy one. See Asahi Metal Indus. Co., Ltd. v. Superior Court, 480 U.S. 102, 107 S.Ct. 1026, 94 L.Ed.2d 92 (1987). “This Court has recognized, however, that it is almost always inconvenient and costly for a party to litigate in a foreign jurisdiction.” Nowak, 94 F.3d at 718. Thus, for this factor to be significant, “the defendant must demonstrate that exercise of jurisdiction in the present circumstances is onerous in a special, unusual, or other constitutionally significant way.” Id.; see also Pritzker v. Yari, 42 F.3d 53, 64 (1st Cir.1994). There is nothing to suggest an especially onerous burden here.
Moreover, as the Supreme Court said in Asahi, “often the interests of the plaintiff and the forum in the exercise of jurisdiction will justify even the serious burdens placed on [an] alien defendant.” 480 U.S. at 114, 107 S.Ct. 1026. In Pleasant St. II, we found that the burden of requiring a Scottish corporate defendant to appear in Massachusetts was “substantially outweighed by Massachusetts’ interest in adjudicating this dispute and plaintiffs’ interest in obtaining convenient and effective relief.” United Elec., Radio & Mach. Workers v. 163 Pleasant St. Corp., 987 F.2d 39, 46 (1st Cir.1993) (Pleasant St. II). This case is similar.
Our cases recognize that courts “must accord deference to the plaintiffs choice of forum.” Nowak, 94 F.3d at 718. As in *636Nowak, a suit involving a Hong Kong defendant, “it is obvious that a Massachusetts forum is more convenient” than a forum in Antigua. Id. Moreover, the United States clearly has a strong interest in the enforcement of its forfeiture laws. The judicial system’s interest in obtaining the most effective resolution of the controversy “also favors the retention of jurisdiction over this dispute.” Pleasant St. II, 987 F.2d at 46. The district court has an interest in ensuring that its own forfeiture order is satisfied and in litigating all claims arising out of Fitzgerald’s criminal proceeding in Massachusetts. See Keeton, 465 U.S. at 777, 104 S.Ct. 1473 (explaining that the forum has an interest in litigating all claims arising out of the underlying libel case).
In discussing the final gestalt factor relating to sovereignty, the Supreme Court has said that when the defendant is a foreign entity, the sovereignty factor of the reasonableness analysis “calls for a court to consider the procedural and substantive policies of other nations whose interests are affected by the assertion of jurisdiction” by the court. Asahi, 480 U.S. at 115, 107 S.Ct. 1026. In this case, there is a potentially significant sovereignty issue that the district court did not reach, and which the parties do not discuss in their briefs. In a January 29, 1998 letter, the Antiguan Cabinet Secretary informed the United States that it froze Fitzgerald’s funds in 1990 “because of evidence that the monies were the proceeds of illegal conduct.” The letter also states: “In a judgement handed down from our High Court dated December 20, 1990, it was found that Fitzgerald was not the owner of these funds.” The letter then says that the Antiguan government discussed the disposition of the funds with SAB after the Massachusetts district court issued its forfeiture order, and that “acting in the public interest of Antigua and Barbuda,” the Antiguan government “released the freeze order on the funds and approved the disposition of the funds in a manner agreed by the Banks and approved by the Government.”
While SAB and IMB, not the government of Antigua, are the defendants in this case, the fifth gestalt factor requires us to take into account the sovereignty concerns raised by this letter. The Antiguan government has claimed $5 million of the forfeited funds, and cites in support of its decision to do so a 1990 order of the Antiguan High Court. Although that claim does not affect the $2 million allegedly converted by SAB, it is an important consideration for the remaining $5 million. Therefore, at least without further briefing by the parties on these sovereignty concerns, I cannot conclude that the government’s showing under the gestalt factors is strong enough to “tip the constitutional balance” here. Nowak, 94 F.3d at 717.10 Nevertheless, the consideration of these gestalt factors reinforces my conclusion that the government’s case for specific jurisdiction was colorable enough to merit the jurisdictional discovery denied by the district court. In my view, that denial was plainly wrong, and an abuse of discretion.
II.
The district court briskly denied the government’s request for jurisdictional discovery, explaining that the government’s showing was “so bootless ... that it has made no colorable claim sufficient to entitle it to any further discovery.” United States v. Swiss Am. Bank, Ltd., 116 F.Supp.2d 217, 225 (D.Mass.2000) (Swiss *637III). That determination is based on a legal misunderstanding of the import of Calder, and therefore constitutes an abuse of discretion. Koon v. United States, 518 U.S. 81, 100, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996) (“A district court by definition abuses its discretion when it makes an error of law.”).
We have held consistently to the rule that a plaintiff may take jurisdictional discovery if its claim is “colorable.” Sunview Condo. Ass’n v. Flexel Int’l, Ltd., 116 F.3d 962, 964 (1st Cir.1997). The “colorable” or “not frivolous” standard for obtaining jurisdictional discovery requires some showing that discovery is needed or likely to be useful. However, that showing is significantly lower than the prima facie showing of jurisdiction, which requires the plaintiff “to demonstrate the existence of every fact required to satisfy both the forum’s long-arm statute and the Due Process Clause of the Constitution.”11 Pleasant St. II, 987 F.2d at 44 (internal quotation marks omitted). The jurisdictional discovery question, by contrast, is whether the government’s showing of minimum contacts falls so far short that discovery is “unnecessary (or, at least, is unlikely to be useful) in regard to establishing the essential jurisdictional facts.” Dynamic Image Techs., Inc. v. United States, 221 F.3d 34, 38 (1st Cir.2000).
Our approach to jurisdictional discovery originates with Surpitski v. Hughes-Keenan Corp., 362 F.2d 254, 255-56 (1st Cir.1966). In that case, we held that the district court should have allowed discovery before ruling on a motion to dismiss for lack of personal jurisdiction where the plaintiff “had at least made good headway, and shown his position not to be frivolous.” Id. at 255. While Surpitski is an older case, we have cited and reaffirmed its discovery-friendly holding numerous times. See Swiss II, 191 F.3d at 46; Sunview Condo., 116 F.3d at 964; Pleasant St. II, 987 F.2d at 48 n. 18; Boit v. Gar-Tec Prods., Inc., 967 F.2d 671, 681 (1st Cir.1992); Whittaker Corp. v. United Aircraft Corp., 482 F.2d 1079, 1086 (1st Cir.1973). In Sunview Condo, we explained that “a diligent plaintiff who sues an out-of-state corporation and who makes out a colorable ease for the existence of in personam jurisdiction may well be entitled to a modicum of jurisdictional discovery if the corporation interposes a jurisdictional defense.” 116 F.3d at 964. Jurisdictional discovery is appropriate “where the plaintiff had been diligent and was somewhat unfamiliar with his adversary’s business practices,” Boit, 967 F.2d at 681, and “where complex factual matters are in question,” Whittaker Corp., 482 F.2d at 1086.
Other circuits similarly allow for discovery when a diligent plaintiff with a color-able but undeveloped case requests it. See Edmond v. United States Postal Serv. Gen. Counsel, 949 F.2d 415, 425 (D.C.Cir.1991) (“As a general matter, discovery under the Federal Rules of Civil Procedure should be freely permitted, and this is no less true when discovery is directed to personal jurisdiction.”); Butcher’s Union Local No. 498 v. SDC Inv., Inc., 788 F.2d 535, 540 (9th Cir.1986) (“Discovery should ordinarily be granted where pertinent facts bearing on the question of jurisdiction are controverted or where a more satisfactory showing of the facts is necessary.” (internal quotation marks omitted)); Compagnie des Bauxites de Guinee v. L’Union Atlantique S.A. d’Assurances, *638723 F.2d 357, 362 (3d Cir.1983) (“Where the plaintiffs claim is not clearly frivolous, the district court should ordinarily allow discovery on jurisdiction in order to aid the plaintiff in discharging that burden.”); Wyatt v. Kaplan, 686 F.2d 276, 283 (5th Cir.1982) (“In an appropriate case, we will not hesitate to reverse a dismissal for lack of personal jurisdiction, on the ground that the plaintiff was improperly denied discovery.”); see also 5A Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure § 1351 at 256-59 (2d ed. 1990) (“In particularly complex cases, ... it may be desirable to hold in abeyance a decision on a motion to dismiss for lack of personal jurisdiction. Doing so will enable the parties to employ discovery on the jurisdictional issue, which might lead to a more accurate judgment than one made solely on the basis of affidavits.”). In sum, “[n]u-merous cases have sustained the right of plaintiffs to conduct discovery before the district court dismisses for lack of personal jurisdiction.” Renner v. Lanard Toys Ltd., 33 F.3d 277, 283 (3d Cir.1994).12
In light of this right, several appellate courts have found, as we did in Surpitski, that district courts erred in denying discovery in cases in which plaintiffs did not allege sufficient facts to make a prima facie case for personal jurisdiction. In Renner, for example, the Third Circuit concluded that discovery should have been granted where the record was “ambiguous” and “incomplete.” Id. at 283. In Edmond, the lower court’s decision to deny discovery was error because the plaintiffs’ allegations were “far from conclusory.” 949 F.2d at 425. In Skidmore v. Syntex Labs., Inc., 529 F.2d 1244 (5th Cir.1976), the court said that discovery should have been allowed because the plaintiffs attorney was not at fault for having failed to discover the requisite jurisdictional facts earlier. Id. at 1248.
Here, the district court based its discretionary denial of discovery on an error of law — its failure to recognize the import of Calder and the need to evaluate more fully the government’s case for jurisdiction. See Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 86 (1st Cir.1998) (mistaken application of law constitutes abuse of discretion); United States v. Snyder, 136 F.3d 65, 67 (1st Cir.1998) (per se abuse of discretion occurs when district court commits error of law). Assessed properly, the government’s case is colorable. As I have explained, Calder held that the in-forum effects of intentionally tortious conduct are a significant jurisdictional contact in their own right. Therefore, when viewed through the prism of the effects test that Calder endorsed, the government’s tort claims are related to SAB’s contacts with the forum. It is under the purposeful availment prong— which the district court never even considered — that the government’s showing falls short. Because Antigua was the legal si-tus of the government’s injury, it is not immediately obvious that SAB expressly aimed its tortious activity at the United States as a forum. Thus, in order to establish a prima facie case, the government cannot rely solely on the in-forum effects of SAB’s actions; it must demonstrate the existence of other contacts between SAB and the forum so that the exercise of jurisdiction over SAB is fundamentally fair.
The government points out that its ability to show more contacts between SAB and the United States, under either a general or specific theory of personal jurisdiction, has been hampered by the bank’s *639privately held status and by Antigua’s banking secrecy laws. Accordingly, the government’s failure to establish the necessary contacts does not necessarily indicate that those contacts do not exist. Rather, it may mean simply that the government has not been able to learn of them without the benefit of discovery. For example, the business contacts between SAB and American companies suggest that there may be more such contacts that the government might be able to discover if it had access to the bank’s records. Similarly, with the benefit of discovery, the government might find out that SAB sent letters or made phone calls to Fitzgerald in the United States, or even sent representatives to meet with him here. Indeed, the government’s investigator has already found phone records indicating that Her-rington placed calls to Boston during the period in which Fitzgerald was setting up his SAB accounts. If the government had access to the bank’s records, it might be able to show that Fitzgerald received those calls, thereby strengthening both the relatedness and purposeful availment elements of its case for specific jurisdiction.
Our precedent in Pleasant St. II is instructive here. The proceedings that led to that decision began when the district court entered an injunction and a contempt order against a Scottish corporation. See 987 F.2d at 42. During the pendency of the corporation’s appeal, the plaintiff proceeded with discovery, but because of the timing of the filings, the discovered material was not part of the record on appeal. Id. We thus vacated the injunction and contempt order for lack of personal jurisdiction in Pleasant St. I unaware of the jurisdictional contacts that the plaintiff had discovered. United Elec., Radio & Mach. Workers v. 163 Pleasant St. Corp., 960 F.2d 1080 (1st Cir.1992). On remand, the district court granted the defendant’s motion to dismiss. The plaintiffs appealed for a second time, and in Pleasant St. II we reversed the dismissal in light of the new facts learned through discovery. We explained:
Under the facts of this case, the incomplete nature of the record prevented any sort of conclusive determination on the personal jurisdiction issue at the time 163 Pleasant St. I was handed down. The jurisdictional deficiency which informed the holding in our previous opinion did not stem from either a settled factual predicate or legally insufficient allegations, but from perceived voids in the evidentiary landscape.
Id. at 47. Noting that before Pleasant St. I, “no discovery directed at filling those voids took place,” id., we continued:
if, on the record before it, the district court had decided the personal jurisdiction issue adversely to plaintiffs without at least affording them the opportunity to ... request discovery, we almost certainly would have declined to affirm the district court’s judgment and held the ruling to be an abuse of the court’s discretion.
Id. at 48 n. 18. In this case, as in the Pleasant St. litigation, the “incomplete nature of the record” rather than a “settled factual predicate or legally insufficient allegations” is the reason that the government cannot make out a prima facie case for jurisdiction. Id. at 47.
SAB launches two additional attacks on the government’s contention that it is entitled to discovery. First, it argues that the care that a court must show in extending its authority over foreign nationals weighs against allowing the government to take discovery. Two circuits, have taken this consideration into account in declining to reverse lower court decisions to disallow discovery. See Cent. States, S.E. and S.W. Areas Pension Fund v. Reimer Express World Corp., 230 F.3d, 934, 947 (7th Cir.2000) (“[I]mposing such burdensome, wide-ranging discovery against defendants from *640a foreign nation is not appropriate at a stage where the district court is trying to determine whether it has any power over the defendants.”); Jazini v. Nissan Motor Co., 148 F.3d 181, 185-86 (2d Cir.1998) (declining to allow plaintiff who made “con-clusory non-fact-specific jurisdictional allegations” to obtain discovery because to do so “would require the federal courts to conduct substantial jurisdictional discovery over foreign corporations — a practice in which they have not hitherto engaged”).
Weighing sovereignty concerns when the plaintiff has not yet shown that the exercise of jurisdiction is proper is indeed a delicate matter. We have urged courts to “exercise even greater care before exercising personal jurisdiction over foreign nationals.” Noonan, 135 F.3d at 93. But our caution does not extend so far as to prevent discovery in a case such as this, where discovery is the only means of filling in the missing pieces of a jurisdictional showing that is more than “colorable.”
The bank argues further that the government has not been a “diligent” plaintiff, as Surpitski and later cases define the term, because it failed to (1) adequately pursue the contacts that it was authorized to investigate pursuant to an Asset Discovery Order issued in the criminal case against Fitzgerald; (2) make adequate use of its interviews with Fitzgerald and Her-rington; and (3) present the district court with a rationale for why discovery would further its case.
The Asset Discovery Order was issued under statutes that authorize discovery “to facilitate the identification and location of property declared forfeited.” 21 U.S.C. § 853(m). SAB argues that the investigation undertaken pursuant to the Asset Discovery Order is the equivalent of discovery.13 However, that Order only authorized discovery on the location of forfeitable assets. It was not a broad discovery tool. The Asset Discovery Order did not give the government access to SAB’s records, which would appear to be the most obvious and promising source of information for the in-forum contacts the government needs to uncover. The government’s investigation to this point has been hampered by its inability to explore these records, an obstacle that court-ordered discovery may (or may not) be able to remove.14
SAB also argues that the government had ample opportunity during the course of its investigations in its earlier prosecution of Fitzgerald and in the present case to obtain information relevant to SAB’s forum contacts. Herrington, SAB points out, was interviewed at length by United States law enforcement officials in 1991 on the Isle of Man, and again by a government investigator after the initiation of proceedings against SAB. Fitzgerald, who had signed a plea and cooperation agreement with the government, presumably was available to provide information relevant to the jurisdictional issues. Given its access to such information, SAB contends, the government already has (or should have) discovered any contacts between SAB and the United States.
That argument is weakened significantly by the fact that Fitzgerald died shortly after the forfeiture order was entered in 1994 — before the events leading to the *641present controversy with the bank — and thus hardly could have aided the government in its attempts to uncover SAJB’s forum contacts. Herrington’s 1991 interview likewise predated the forfeiture order and SAB’s failure to comply. Accordingly, the government had no reason to press him regarding his or SAB’s contacts with the United States.15 Rather, the interview focused on facts relevant to the criminal charges of conspiracy and money laundering that later were brought against Fitzgerald and several other individuals. A government investigator did conduct a brief telephone interview with Herrington in 1998, after the government filed its complaint in the present action. But the apparent purpose of the interview was to gather information demonstrating IMB’s control of SAB, not to determine the extent of the latter’s forum contacts. In any event, that interview does not alter the government’s status as a “stranger” to SAB within the meaning of Surpitski, 362 F.2d at 255, and its progeny. See, e.g., Whittaker Corp., 482 F.2d at 1086 (noting that jurisdictional discovery is appropriate where, inter alia, a party is “somewhat unfamiliar with his adversary”); Am. Express Int’l, Inc. v. Mendez-Capellan, 889 F.2d 1175, 1181 (1st Cir.1989) (finding that parties were not “total stranger[s]” under Surpitski where they had a “long commercial relationship”). Indeed, the government’s attempt to investigate only underscores that its relationship with the bank is an artifact of the forfeiture order. While Fitzgerald had business dealings with SAB, and so was not a stranger to the bank, the government had no such ongoing relationship.
Finally, SAB argues that the government did not meet its burden of explaining to the district court the discovery sought and its value. We have said that plaintiffs must “explain! ] ... how discovery, if allowed, would bear on the narrow jurisdictional issue.” Dynamic Image, 221 F.3d at 39. In opposing SAB’s motion to dismiss, the government articulated the theories of general and specific jurisdiction that it was trying to prove and requested discovery of “any information regarding the existence, nature and scope of SAB contacts with the United States and United States persons.” As the majority points out, only on appeal did the government fully explain the types of contacts it hopes to discover. The majority is correct to disregard specifics not presented below. In my view, however, the government adequately explained to the district court the purpose of its request for discovery, and its description of the contacts it hoped to find, while bare, meets the diligence standard.16 After all, it is obvious that the *642government seeks evidence of physical, telephone, or mail contacts that are lacking in the current record.
In short, the government was a diligent plaintiff with a colorable claim. See Surpitski, 362 F.2d at 255; Sunview Condo., 116 F.3d at 965. If given the opportunity for appropriate discovery, it may well be able to make out a prima facie showing of specific jurisdiction. The district court did not recognize that possibility, however, because it refused to treat the in-forum effects of SAB’s allegedly tortious activities as a jurisdictional contact. As a result, the court ended its specific jurisdictional analysis with the relatedness element, and summarily denied the government’s request for discovery. Based as it was on a mistaken application of Colder, that denial was “plainly wrong,” Crocker v. Hilton Int’l Barb., Ltd., 976 F.2d 797, 801 (1st Cir.1992).17 See Ruiz-Troche, 161 F.3d at 86. Moreover, the denial caused the government “substantial prejudice.” Crocker, 976 F.2d at 801. Without discovery, the government’s case ends.
III.
Because I conclude that the district court erred in refusing to allow jurisdictional discovery with respect to the government’s claims against SAB, I would also vacate the dismissal of the case against IMB. The district court determined that the government had failed adequately to plead alter ego liability against IMB, and that it had not established a sufficient basis for personal jurisdiction. We said in Swiss II that any ruling on alter ego liability was “premature,” because the jurisdictional question should be resolved before reaching the merits of the case. 191 F.3d at 46 (citing Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 118, 118 S.Ct. 1003, 140 L.Ed.2d 210 (1998); Ruhrgas AG v. Marathon Oil Co., 526 U.S. 574, 119 S.Ct. 1563, 143 L.Ed.2d 760 (1999)). The factors discussed in Swiss II continue to weigh in favor of that approach here.
As the majority explains, personal jurisdiction over IMB is contingent on the government’s ability to make out a prima facie case for jurisdiction over SAB. The district court ruled on the latter question without having allowed discovery against SAB to proceed. I would remand the case so that such discovery could take place. If the government, with the benefit of jurisdictional discovery, were able to establish a prima facie case of jurisdiction over SAB, the district court would then have to reassess the jurisdictional status of IMB and *643its alter ego ruling, as well as any discovery issues relating to IMB.
. Calder clarified that the purposeful availment requirement is met whenever the defendant intentionally reaches out to the forum in some way, whether it is seeking benefits or causing harm. The Court reaffirmed that point in Burger King Corp. v. Rudzewicz, explaining that due process requires that individuals have "fair warning" that their activities might subject them to jurisdiction in the forum, and that the fair warning requirement is satisfied if the defendant " 'purposefully directed’ his activities at residents of the forum and the litigation results from alleged injuries that 'arise out of or relate to’ those activities.' ” 471 U.S. 462, 472, 105 S.Ct. 2174, 85 L.Ed.2d 528 (1985) (quoting Keeton v. Hustler Mag., Inc., 465 U.S. 770, 774, 104 S.Ct. 1473, 79 L.Ed.2d 790 (1984); Helicopteros, 466 U.S. at 414 n. 8, 104 S.Ct. 1868).
. We reasoned in Kowalski and Sawtelle that the effects of an injury are not the same thing as the injury itself. See Kowalski, 787 F.2d at 11; Sawtelle, 70 F.3d at 1390. I address the distinction between injury and effects below, as part of the purposeful availment analysis.
. It bears emphasis that the weighing analysis should be done in the first instance by the district court, which should not have ended its specific jurisdiction inquiry with the relatedness element.
. The government argues that our admonition in Swiss II, 191 F.3d at 45, that "[a] timely and properly supported request for jurisdictional discovery merits solicitous attention,” further softens the "colorable” standard. That is not so. Rather, the "timely” and "properly supported” language reflects our statements elsewhere that a plaintiff must be "diligent” to merit discovery. See, e.g., Sunview Condo., 116 F.3d at 964.
. But see Jazini v. Nissan Motor Co., 148 F.3d 181, 186 (2d Cir.1998) ("Since the Jazin-is did not establish a prima facie case that the district court had jurisdiction over Nissan Japan, the district court did not err in denying discovery on that issue.”).
. SAB makes much of the district court’s statement that the government was not entitled "to any further discovery.” Swiss III, 116 F.Supp.2d at 225. I assume that the court’s use of the word "further” simply refers to the government's investigation pursuant to the Asset Discovery Order.
. In its November 13, 1995 letter to the government, SAB said that the relevant records were destroyed in a hurricane. The government presumably would test this assertion if it were permitted to pursue jurisdictional discovery.
. During that interview, Herrington was questioned about certain conversations with Fitzgerald in which he explained how Fitzgerald's anticipated deposits would be handled by SAB. Herrington indicated that all those conversations took place in Antigua. When asked whether “they all were face to face,” he answered, "I, in the best of my knowledge, er, I never met Mr. Fitzgerald anywhere else but Antigua.” The government interviewer did not ask Herrington whether he ever had contacted Fitzgerald by other means (for example by mail or telephone). There was no cause for the government to seek such details in the context of its 1991 investigation.
. SAB also faults the government for not renewing its motion for discovery before the bank filed its motion to dismiss following remand. The timing of the government’s motion was proper. The Federal Rules of Civil Procedure do not provide an opposing party an explicit right to discovery in the motion to dismiss context, and the government could best explain to the court why it merited discovery in response to the arguments in SAB’s motion to dismiss. The government preserved its request for discovery at each juncture of this case, in contrast to plaintiffs in other cases in which we have affirmed denials of requests for discovery. See Dynamic Image, 221 F.3d at 38; Sunview Condo., 116 F.3d at 964; Boit, 967 F.2d at 681.
. Contrary to the majority’s suggestion, Crocker does not stand for the proposition that the district court retains "broad discretion to decide whether discovery is required” even if the plaintiff has been diligent and has made a colorable claim for personal jurisdiction. Crocker, 976 F.2d at 801. In affirming the district court’s denial of discovery in Crocker, we did not so much as hint drat the plaintiffs' case was colorable, or that they had been diligent. Instead, we simply observed that discovery would have been futile, as the information the plaintiffs sought would not have established that the defendant did business in Massachusetts, as required by that state’s long arm statute. See id. In so doing, we noted the district court’s broad discretion in considering such questions, and explained that its decision would be overturned " 'only upon a clear showing of manifest injustice, that is, where the lower court’s discovery order was plainly wrong and resulted in substantial prejudice to the aggrieved party.' ” Id. (quoting Santiago v. Fenton, 891 F.2d 373, 379 (1st Cir.1989) (discussing standard for pre-trial, non-jurisdictional, discovery)). Under our precedents dating back to Surpitski, a district court's order would be "plainly wrong” if, without any reason to the contrary, it denied jurisdictional discovery to a diligent plaintiff with a colorable claim.